How much behavioral information can DNA store?

[CaveDave]

It would not be enough to be able to learn new behaviors; one would have to be able to distinguish between behaviors one should learn and ones which one should not. For instance, if you see a bird jump out of a tree, that is a behavior that you shouldn't learn.

Really? I can imagine many circumstances where the ability to outwit your enemy by exhibiting a completely unexpected response might allow you to live long enough to reproduce, while a programmed response could end in your demise because it was predictable. Maybe the jumping bird would survive the lizzard attack on it's nest while the ones that screeched at the attacker were eaten?

[B And hasn't this beem accomplished, in part, by including more hard wired subroutines? [/B]

Well, that still limits them to the set of anticipated needs, not allowing for much beyond that.

Just my opinions.

Dave

Edited for spelling

 

[Art Vandelay]

I can imagine many circumstances where the ability to outwit your enemy by exhibiting a completely unexpected response might allow you to live long enough to reproduce, while a programmed response could end in your demise because it was predictable.

My point is that a randomly selected "unexpected response" is, on average, going to be worse than just instinct. Being able to figure out which new response to employ requires advanced instinct.

Maybe the jumping bird would survive the lizzard attack on it's nest while the ones that screeched at the attacker were eaten?

The bird jumping and expecting to fly would do better. The human jumping and expecting to fly is going to die.

Well, that still limits them to the set of anticipated needs, not allowing for much beyond that.

Every learned behavior can be a liability given the right situation. No Free Lunch Theorem.

 

[CaveDave]

My point is that a randomly selected "unexpected response" is, on average, going to be worse than just instinct. Being able to figure out which new response to employ requires advanced instinct.

It does? I don't think I mentioned random, just unexpected to the opponent: this could involve careful advance planning and strategy based on moves the opponent wouldn't expect from previous experience. I understand that man's ability to overcome many beasts is based on our actions not being in line with the animal's wired expectations (such as ambush/coralling/projected weapons/etc)

The bird jumping and expecting to fly would do better. The human jumping and expecting to fly is going to die.

Doesn't it sort of depend on the altitude and what is below to break the fall? What human would expect to fly unless perhaps they had made advance preparations for this escape?

Every learned behavior can be a liability given the right situation. No Free Lunch Theorem.

Sure. But in the same regard, can't any wired-instinctual behavoir be just as lethal? My assertion is that among a population faced with a given chalenge, the ones with a flexible response might have a better chance of some members surviving to reproduce, while those with a fixed reaction might get to die together.

Dig a circular trench, install a bridge and a narrowing barricade, drive elephants toward the strait, remove bridge. Elephants are captured: they won't escape because their trunks can't feel the trench bottom. Their limited adaptability, although usefull in normal conflicts with other beasts, helps them not at all against human deviousness.

Dave

 

[CBL4]

I don't think I mentioned random, just unexpected to the opponent: this could involve careful advance planning and strategy based on moves the opponent wouldn't expect from previous experience.

The problem with this is that if it is successful, the gene will become more and more prevalent. Once it is prevalent, the opponent will start to expect it and it will no longer be advantageous.

Game theory would probably show that it would be most likely to survive in a small percentage of the population. However, if the population is decimated, it could easily die out.

CBL

 

[CBL4]

Have anyone determined how the genes cause the innate behaviors?

Presumably some protein(s) are formed or not formed at some particular time. My real question is whether it is during development or at the time of the behavior.

CBL

 

[CaveDave]

The problem with this is that if it is successful, the gene will become more and more prevalent. Once it is prevalent, the opponent will start to expect it and it will no longer be advantageous.

True, but I was meaning non-genetic, learned (and modifiable) behavior.
Do cats or birds ever learn to ignore the detatched tail and chase the front of the lizzard? (Although, I did once have an old tomcat that got that way by learning to stop and look when crossing roads. )

Game theory would probably show that it would be most likely to survive in a small percentage of the population. However, if the population is decimated, it could easily die out.

Or, it could be made more viable by the weeding-out of those less adaptable members.

Just my thoughts.

Dave